Mind2: Mind-to-Mind Emotional Support System with Bidirectional Cognitive Discourse Analysis

TL;DR

Mind2 presents a bidirectional cognitive discourse framework for emotional support dialogue generation, leveraging Theory-of-Mind, psychological expected utility, and cognitive rationality to create interpretable, context-aware ES conversations. It introduces a dynamic discourse context propagation window and a prompt-based BCK synthesis pipeline, enabling system- and user-centered reasoning within evolving dialogues. Empirical results on the ESConv dataset show Mind2 achieves competitive performance with only 10% of training data and further gains with more data, with ablations confirming the value of each cognitive component. The approach offers a path toward more transparent, user-tailored ES systems and highlights avenues for improved local context control and domain adaptation.

Abstract

Emotional support (ES) systems alleviate users' mental distress by generating strategic supportive dialogues based on diverse user situations. However, ES systems are limited in their ability to generate effective ES dialogues that include timely context and interpretability, hindering them from earning public trust. Driven by cognitive models, we propose Mind-to-Mind (Mind2), an ES framework that approaches interpretable ES context modeling for the ES dialogue generation task from a discourse analysis perspective. Specifically, we perform cognitive discourse analysis on ES dialogues according to our dynamic discourse context propagation window, which accommodates evolving context as the conversation between the ES system and user progresses. To enhance interpretability, Mind2 prioritizes details that reflect each speaker's belief about the other speaker with bidirectionality, integrating Theory-of-Mind, physiological expected utility, and cognitive rationality to extract cognitive knowledge from ES conversations. Experimental results support that Mind2 achieves competitive performance versus state-of-the-art ES systems while trained with only 10\% of the available training data.

Figures (2)

• Figure 1: An overview of the Mind2 framework.
• Figure 2: An example of Mind2 ES dialogue generation. In contrast to the baseline, Mind2 model enhanced with three BCK components produces ES dialogues that are contextually more congruent with the target responses.